Time-Series Clustering

This project focuses on clustering time-series data using various techniques, including derivatives, Fourier transformations, and Dynamic Time Warping (DTW).

Table of Contents

• Overview
• Techniques Used
• Installation
• Citation

Overview

Time-series clustering is a method used to group similar time-series data based on specific characteristics or patterns. This project implements a clustering techniques using derivatives, Fourier transformations, and DTW to analyze and group time-series data effectively.

Techniques Used

• Derivative-Based Clustering: Analyzes the rate of change in time-series data to identify patterns.
• Fourier Transformation: Converts time-series data into the frequency domain to detect periodic patterns.
• Dynamic Time Warping (DTW): Measures similarity between time-series sequences that may vary in time or speed.

Installation

To set up the project environment, follow these steps:

1. Clone the repository:

   git clone https://github.com/zxnga/TS-Clustering.git
   cd TS-Clustering

2. Create a virtual environment (optional but recommended):

   python3 -m venv env
   source env/bin/activate

3. Install the required dependencies:

   pip install -r requirements.txt

Citation

If this project contributes to your research or work, please cite the following paper:

@article{zangato2025data,
  title={Data-driven policy mapping for safe RL-based energy management systems},
  author={Zangato, Th{\'e}o and Osmani, Aomar and Alizadeh, Pegah},
  journal={Energy Reports},
  volume={13},
  pages={1888--1909},
  year={2025},
  publisher={Elsevier}
}